Professor Peter Woodward of Heriot-Watt University in Edinburgh said there is a risk the lines may not be able to cope with "new problems" thrown up by the higher speeds with times between London and Birmingham cut to 49 minutes.
The UK Government has reportedly given £500,000 to Mr Woodward – professor of railway geo-engineeering – and Professor Mike Forde, of Edinburgh University, to investigate the issue.
However, in papers lodged with the Government's Engineering and Physical Sciences Research Council, Mr Woodward said the speeds proposed for the £24 billion HS2 project could trigger "significant amplification of train-track vibrations".
It added this may cause "rapid deterioration of the track, ballast and sub-ballast, including possible derailment and ground failure".
HS2's two-track core route between London and the Midlands is supposed to handle traffic currently served by three separate main lines, to destinations including Birmingham, Manchester, the East Midlands, Sheffield, Leeds and Scotland.
Business leaders say journey times between London and Edinburgh and Glasgow would be cut by 30 minutes from four hours and 30 minutes. But experts are worried about plans to run trains faster than any other line in the world, 225mph initially AND with a target speed is 250mph within a few years of opening in 2032.
Mr Woodward told The Engineer magazine he was worried about "ground waves" of vibration, known as Rayleigh waves, developing in the rail at a certain speed. "The analogy is that of an aircraft going through the sound barrier," he said.
As "critical track velocity" was approached, the track would "start to undergo strong ground vibrations", rippling visibly along the rails in front of the train.
"It is possible that, if the train was allowed to run at this critical track velocity, it would derail at high speed," he warned.
Using a test track bed, Mr Woodward is trying to find out what the danger speed is and whether HS2 will exceed it.
Emails released under Freedom of Information show Andy Went, HS2's head of track, now a senior engineer at Network Rail responsible for HS2, is closely involved in the research. They also show Hs2 chief engineer Professor Andrew McNaughton was closely involved in drawing up the new research proposal.
This HS2 involvement apparently calls into question Mr McNaughton's assurance to the Transport Select Committee that a 225mph service can run safely on current forms of track.
"We do not consider it requires technology development to achieve [the proposed service] at 225mph and we believe only limited, foreseeable development would be necessary for [250mph]," he said.
HS2 claims it will be able to run up to 18 trains an hour – one every 3min 20sec – along the core route, more than on any other such high-speed line in the world. If the trains were slower, frequencies would have to be reduced, putting at risk the promised service. Running at the industry standard of 186mph would cut scheme's benefit-cost ratio by an estimated 15%.
A spokesman for HS2 said: "It is nonsense to suggest we would design a railway that did not take into account the effects of Rayleigh waves. We are designing HS2 so that there is no possibility Rayleigh waves would lead to any problems such as derailment.
"We will continue to work closely with and support leading researchers such as Professor Woodward to develop and design safe and efficient high-speed train operations in the UK."
Mr Woodward could not be reached for comment yesterday.